Respiratory syncytial virus (RSV) can be a very devastating nosocomial infection within the confines of the neonatal intensive care unit. Precautions to protect other low birth weight infants must be taken when this infection is present. The diagnosis of this infection however, its based upon antigen detection using enzyme immunoassay (EIA) or culture. We postulated that the EIA test does not work well in this population. The purpose of this study was to evaluate the sensitivity of RSV antigen detection to culture using EIA in low birth weight infants. Records of all RSV EIA tests ordered from the neonatal intensive care unit from the winters of 1990-1991 thru 1996-1997 were obtained. Patients were eligible for inclusion if they weighed less than 2000 grams at the time of evaluation. Patient demographic data and the source of the specimen (TA vs NP) were gathered. RSV EIA and RSV culture results were obtained and the sensitivity of this laboratory examination was calculated. Eighty-seven patients were identified while 29 were > 2000 grams and 6 patient records could not be found. Fifty-two patients were eligible for the study.These 52 patients underwent 68 evaluations for RSV. This cohort of patients had a mean birth weight and gestational age of 978 grams(range:590-1972) and 27 weeks (range:23-36). All patients were evaluated because of the development of respiratory symptoms at a mean age of 36 days(range:2-148). The mean weight at the time of evaluation was 1253 grams(range:650-1964). Fifty had negative EIA and culture (25-NP), 7 had positive EIA (5-NP), and 11(5-NP) had negative EIA and positive cultures. The sensitivity of the EIA was 39%. The sensitivity of the EIA for all specimens submitted to our laboratory from the winters of 1992-1993 thru 1996-1997 was 75% (p < 0.01). The sensitivity of the RSV EIA was significantly lower in low birth weight neonates as compared to older children. Care should be taken in interpreting negative RSV EIA results in low birth weight patients.